一种改进的NAS-RIF图像盲复原算法  

An Improved NAS-RIF Algorithm on Image Blind Restoration

作　　者：曲振峰[1] 

机构地区：[1]洛阳理工学院电气工程与自动化系,河南洛阳471023

出　　处：《洛阳理工学院学报（自然科学版）》2010年第2期48-51,55,共5页Journal of Luoyang Institute of Science and Technology：Natural Science Edition

摘　　要：在重点研究了具有非负和有限支持域的递归逆滤波器的盲图像复原算法(即NAS-RIF算法)的基础上,研究了一种通过高阶统计量HOS去除噪声的方法,抑制了高斯噪声,保留了更多的图像细节,并在算法的迭代过程中采用图像分割技术进一步精确图像的目标支持域,提高了该算法的应用范围。由仿真实验的结果可以看出,改进后的方法取得了比较好的复原效果。Based on the NAS-RIF algorithm, a denoising method has been studied through higher order statistics to restrain Gaussian noise, preserve more image details, extend the algorithm application and acquire more accurately defined images＇ targeted supporting domain through image segregation technique in the process of iteration of algorithms. Results of the simulation experiments show that the modified method has achieved better restoration effects.

关 键 词：数字图像处理 图像盲复原 图像去噪 NAS-RIF 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]